446 



Mr. W. F. K. Weldon. The Variations [Apr. 17, 



Mr. Galton lias, as is well known, studied the frequency with which 

 deviations from the average size of certain organs occur in man, in 

 certain plants, and in moths. The result of his investigations has 

 been to show that deviations from the average size of the organs 

 measured by him occur in every case with the frequency indicated by 

 the law of error. These results were, however, based on an investi- 

 gation either of civilised man, or of a domesticated animal or plant : 

 and Mr. Galton has himself pointed out that in the majority of cases 

 studied, the effect of natural selection is probably insignificant. The 

 similar investigations of Quetelet and others are also confined to 

 civilised man. It has, therefore, seemed worth while to attempt an 

 investigation of the variations in the size of certain organs which 

 occur in a species living in a wild state, upon which natural selection 

 and the other destructive or plastic influences from which domestic 

 animals and civilised man are alike protected may be supposed to act 

 with full effect. 



In his recent work on heredity,* Mr. Galton predicted that selec- 

 tion would not have the effect of altering the law which expresses 

 the frequency of occurrence of deviations from the average : so that 

 he expected the frequency, with which deviations from the average 

 size of an organ occurred, to obey the law of error in all cases, 

 whether the animals observed were under the action of natural selec- 

 tion or not. The results of the observations here described are such 

 as to fully justify Mr. Galton's prediction. 



These observations relate entirely to the lengths of organs, or parts 

 of organs. The measurements of these lengths were made either 

 with a pair of compasses, in the case of the greater lengths, or in 

 the case of smaller parts by means of a microscope, provided with 

 cross-wires, and travelling on a screw of known pitch. The results 

 are, I believe, accurate to within about 0*1 mm. The edges of the 

 parts measured were in many cases so uneven, and the effect of the 

 spirit in which the specimens were preserved was probably so con- 

 siderable, that a greater degree of apparent accuracy in the measure- 

 ments would not have implied a more reliable result. 



In order to compare the organs of one individual with the corre- 

 sponding parts of another individual of different size, it was evi- 

 dently necessary to express the dimensions of each organ in terms of 

 the length of the body of the individual to which it belonged. All 

 the measurements used in this paper are therefore, expressed in 

 terms of the total length of the body, which is taken as 1000. 



Having obtained measures of the length of an organ in a suffi- 

 ciently large sample of individuals, the frequency with which the 

 various magnitudes occur may be conveniently exhibited in the fol- 



* 1 Natural Inheritance,' pp. 119 — 124. 



